Model moves Alaska Senate back toward Sullivan, without a new poll to point to

Our model now gives Mary Peltola a 25 in 100 chance of winning the Alaska Senate race, down from 35% in our run of August 17. That is a drop of 10.1 points of Democratic win probability over seven days, all of it toward Republican incumbent Dan Sullivan.

The estimated margin moved with it. Our mean margin in this race is now R+4.5, against R+2.7 a week ago. The spread of simulated outcomes narrowed slightly: our 5th-to-95th percentile range runs from R+15.6 to D+6.2, compared with R+13.9 to D+8.3 last week. A quarter-chance is not a small chance, and the range still contains outcomes on both sides.

We cannot attribute the shift to new survey data. The most recent Alaska poll available to this run is a Data for Progress survey of 575 likely voters, fielded July 28 to August 4, which had Peltola at 50 percent and Sullivan at 45 percent. That poll predates our August 17 run and was already in our average when this site last wrote about the race. Nothing fielded since then appears in this run. The model moved; the cause is not something the inputs here identify, and we are not going to guess at one.

Nationally, our model gives Democrats a 45 in 100 chance of Senate control. Democrats need 51 seats for control and Republicans 50, because the vice president is a Republican and a 50-50 tie goes to the Republicans. Our mean Democratic seat count is 50.0 — right at the line where that tiebreak decides the chamber, which is part of why a single state moving ten points of win probability registers at the national level at all. Election day is November 3, 71 days out.